Considerations in the Development of Vaginal Products Intended to Prevent the Sexual Transmission of HIV

Sheryl L. Lard, Kenneth L. Hastings, and David Feigal*
Division of Antiviral Drug Products, CDER, FDA

*Corresponding author.

Abstract

Vaginal products intended to prevent transmission of the Human Immunodeficiency Virus (HIV) represent a new category of topical agents with important public health applications. Included in this category are a highly diverse group of drugs and biologics that vary not only in terms of chemical composition, but also with respect to their ability to support other indications (e.g., contraception, broad spectrum STD prophylaxis. As a consequence of this heterogeneity vaginal products seeking an HIV prevention claim are reviewed by the FDA on a case-by-case basis, taking into account product characteristics and intended use to guide research and development and to determine the regulatory path to approval.

This chapter will address basic issues in drug development and regulatory considerations that apply to chemical barrier products seeking an HIV prevention indication.

Introduction

Recent estimates suggest that over 75% of the 10-12 million HIV-infected adults worldwide contracted the virus through heterosexual intercourse¹, and it is now the primary route of infection with HIV for women in the United States^2-4. These statistics imply that readily available methods to prevent sexual transmission of this virus (i.e., condoms or abstinence), are not sufficient to meet the diverse needs of all women at risk of acquiring the infection. Male condoms, when used properly, are an effective first line, defense against many sexually transmitted disease (STD) agents, including HIV^5,6. However, this device requires male consent and is also a form of contraception, two factors that limit its utility in the population at large^7,8. Abstinence, while clearly an effective way to avoid STDs, has not been readily accepted as a long-term solution and simply may not be an option for many high risk women. Social, religious and practical considerations suggest that new methods for infectious disease prophylaxis are needed that are female controlled and that provide freedom of choice with respect to procreation^8-10.

General Regulatory Considerations

Vaginal products seeking an HIV prevention claim may travel a unique and often multi-faceted, regulatory path to approval. The primary reason for this multi-disciplinary approach is that vaginal HIV prophylaxis is generally not a "stand-alone" indication. Many potentially useful topical microbicides are active against a variety of different pathogens and hence, may seek a more global STD prevention indication. Furthermore, vaginal prophylaxis products can take a number of different forms (i.e., simple chemical barrier, chemical + physical barrier, immunologic + chemical barrier, etc). Finally, from a regulatory standpoint, potential reproductive effects are an important consideration when evaluating a topical agent intended for vaginal use during sexual intercourse. Each of these factors can influence the type of data required for marketing approval and will dictate which groups at the FDA will participate in the product review process.

In general, FDA review authority for a pharmaceutical agent is determined on the basis of product characteristics and intended use. For example, vaginal microbicides seeking an HIV prevention claim are reviewed by the Division of Antiviral Drugs, within the Center for Drug Evaluation and Research (CDER). The same division also reviews drugs intended to prevent other sexually transmitted diseases of viral origin. However, a general STD prevention claim must address a variety of different organisms in addition to viruses. If a product seeking an HIV prevention indication also claims activity against sexually transmitted bacterial, fungal or parasitic infections, the Division of Anti-infective Drug Products joins in the review process. This simple dichotomy is complicated by the fact that a vaginal product may also prevent conception. The division of Metabolic and Endocrine Drug Products has the responsibility for review of all drug-based contraceptive efficacy claims. In addition, if a product will be labeled specifically for use with a device such as a condom, diaphragm or cervical cap, or is actually part of a device, the Center for Devices and Radio logic Health will also participate in the review. Finally, if the active ingredient was approved for sale "over the counter" (OTC) as is the case for nonoxynol-9 (N-9), monograph standards may also apply. At a minimum, this requires consultation with the Office of OTC Drug Evaluation. These possibilities are all based on the assumption that the product under consideration is a drug or a device plus a drug. It should be noted that candidates for STD prevention and/or vaginal contraception can be biologics (e.g., monoclonal antibodies or recombinant proteins). These products are reviewed by the Center for Biologics Evaluation and Research (CBER).

Rather than leave a potential sponsor with the difficult task of defining exactly where an application should be submitted, the agency maintains a central document facility that assigns investigational products to a review division based on the proposed indication and product characteristics. The review team then takes the responsibility for determining if a consultation with another division is indicated, or if a separate submission to a different division is warranted. If additional submissions are required, the Consumer Safety Officer can assist the sponsor in making contact with the other review division(s).

With regard to contraceptive claims, a proposed rule was published February 3rd, 1995 which addresses the OTC status of N-9 and otoxynol 9-based vaginal contraceptive drug products^11. The FDA has determined that the efficacy of these drugs is highly variable and dependent on the final formulation of each product. Given that in vitro testing does not fully predict the actual effectiveness of these products, the agency has proposed that manufacturers conduct clinical trials and obtain approved product-specific applications for marketing of their products. While the proposed rule does not directly affect vaginal products seeking an STD prevention claim, the contraceptive efficacy trials performed to meet this requirement will provide an opportunity to gain new insights on the role of formulation in product activity and toxicity.

Steps in the Drug Approval Process

While there may be more than the usual number of scientific disciplines involved in the review, the actual process of drug development for a vaginal microbicide is very similar to what is done for most new therapeutic agents. A New Drug Application (NDA) is required for FDA approval to market a vaginal product for HIV prevention. As for any antimicrobial drug, the NDA should include preclinical data on the activity and toxicity of the active ingredient(s) alone and/or in the context of the final formulation, information pertaining to the composition and manufacture of the product and data from at least two adequate and well controlled clinical trials. This description reflects a typical NDA, but the final contents of any given application is usually determined on a case-by-case basis through discussions with FDA reviewers. Given that the NDA generally represents an end product of the drug development process, open communication between the sponsor and the FDA should be established well in advance of NDA discussions. In the Division of Antiviral Drugs at CDER, drug sponsors are encouraged to bring a new product to the attention of the FDA as a "preIND", that is prior to the sponsor's request to initiate clinical studies in humans.

Preclinical activity and toxicity data are typically generated very early in the drug development process. This information then serves as the basis for a preliminary risk/benefit assessment to determine product suitability for Phase I/II clinical testing. When appropriate preclinical studies are completed or well underway, an Investigational New Drug (IND) application is filed and if preclinical evidence of safety is sufficient, Phase I clinical studies can be initiated. Phase I studies are usually conducted in normal individuals and are designed to examine the safety and tolerability of a new product. Dose ranging and pharmacokinetic studies may also be conducted in Phase I trials. Nonclinical drug activity studies should be completed, or well underway before Phase II trials are begun. Phase II clinical trials are small scale efficacy studies intended to provide information on product activity in a limited number of individuals. Consequently, it is important to know as much as possible about the antiviral activity of the product and its mechanism of action prior to the design of these trials. The cumulative data base obtained from preclinical and clinical Phase I/II activity and toxicity studies is then used to design Phase Ill confirmatory studies in humans. These "pivotal" trials are intended to demonstrate product efficacy in a large number of subjects.

 

Important Issues in Vaginal Product Development

Presented below is a brief description of the major components of an NDA submission, mom ding a discussion of some special considerations for the development of vaginal WV prophylaxis products. This is not an all inclusive list of studies to be performed during the preIND and IND stage, but rather an overview of important issues that should be considered during the development of a product for this indication.

Microbiology

The microbiology section of the NDA contains the preclinical activity data used to support human testing, and may include relevant clinical data from ex vivo virologic or immunologic studies. The principle goal of the preclinical microbiology studies is to accurately define the activity of the agent in a nonclinical setting that most closely approximates the conditions under which the product will be expected to perform during human use. A secondary goal of these studies is to provide data that can be used to optimize product performance in the clinic. This may include microbiologic support for formulation studies and stability testing, the identification of product limitations (e.g., time course of kill for virucidal compounds or pretreatment requirements for nonvirucidal blocking agents), or in vitro assessments of the duration of product action. As noted in the previous chapter, there are specific types of preclinical and clinical data that are expected for more traditional forms of prophylactic topical antiseptics (e.g., handwashes, gloving creams, surgical scrubs, or topical antibiotics). Many of the same scientific concerns apply to the study of prophylactic vaginal products, but with the caveat that these preparations must function in a very different environment than that of intact or even abraded skin surfaces.

Preliminary evidence of antiviral activity often provides the impetus to develop a compound for an HIV prevention indication, but is usually not sufficient to support Phase II clinical trials. Demonstrating that a product can kill or block HIV in a screening assay that employs laboratory-adapted virus isolates and continuous cell lines is only the first step in the assessment of activity A dose response curve for activity and cytotoxicity are needed to guide dosage selection for early clinical studies. In view of the type(s) of virus that may be transmitted via semen, activity should to be determined against clinical and laboratory isolates of HIV (both cell-associated and cell-free virus), using peripheral blood mononuclear cells as targets for infection in addition to (or in place of), continuous cell lines. This is particularly important for agents that interact directly with the HIV virion and/or interfere with virus receptor-cell interactions. Preclinical studies have demonstrated that virus strains that grow well in continuous T-cell lines (a characteristic shared by many lab-adapted viruses), differ from monocytotropic viruses in a variety of ways including CD4 binding affinity, susceptibility to antibody neutralization, net positive charge of the V3 loop and use of specific cellular cofactors for viral binding^12-14. Failure to test a candidate compound against clinically relevant virus isolates of both the T-cell tropic and monocytotropic phenotype may suggest that antiviral activity against both phenotypes is present when it is not. Alternatively, exclusive use of lab isolates may skew the dose-response curve, thus interfering with the selection of an appropriate dose of the active ingredient(s) for early clinical efficacy trials.

Another important issue to address during preclinical development is the mechanism of drug action. At a minimum, studies should be conducted to determine if a product inactivates the virus (i.e., is virucidal or neutralizing), or blocks virus-cell interactions with no direct effect on virus infectivity (e.g., agents that alter cells to resist infection or physically block the interaction between the virus envelope and the cell surface). These data can help to identify appropriate preclinical studies to validate activity (e.g., whether animal protection studies are strongly recommended) and may influence formulation decisions for subsequent clinical trials.

In addition to understanding how a product acts and at what concentration in tissue culture, it is important to know how the environment of the vaginal vault will impact on drug activity. A product introduced into the vagina will be diluted to a certain extent by mucosal secretions and again with the introduction of semen. Product activity may be improved by this interaction, unchanged or decreased. For example, a mucospissic agent can turn mucosal secretions into a physical barrier that can trap virus or reduce access to susceptible target cells. Alternatively, a pH sensitive product may be rapidly degraded by the low pH (3.7-4.8) environment of the vagina resulting in product inactivation or a limited duration of activity.

Clearly, timing is also an important issue to consider in the development of a vaginal product that must be present and active during intercourse. The duration of drug action will determine how far in advance a product may be used and still provide protection. This may depend as much on physical characteristics of the formulation (e.g., bioadherence properties), as it does on chemical compatibility of the active ingredient(s) with mucosal secretions. Duration of activity should not be confused with the time required for product activity (sometimes referred to as the "time course of kill" for virucidal products). This is defined as the minimum amount of time required for a product to impart its protective effect (e.g., the time required to kill 4 logs of virus, or the amount of time cells must be pretreated in order to resist infection). The minimum time required for product activity is also an important variable that should be defined prior to the initiation of clinical efficacy studies. Many of the issues discussed above may be investigated in vitro (using secretions obtained from normal or HIV+ donors), or through ex vivo studies conducted with samples obtained during Phase I clinical trials.

Preclinical Toxicology

To support the safety of proposed clinical trials with vaginal products, a product sponsor must conduct preclinical safety studies designed to adequately model in mammals the intended clinical use of the drug. Toxicology studies should be conducted in at least two animal species (one a non-rodent) and should use the intended route of exposure (i.e., vaginal administration) The rabbit is typical[y the non-rodent species of choice for vaginal irritation studies, but is not specifically required. Animal studies should employ the final formulation intended for clinical use in doses that are comparable, based on appropriate methods of interspecies extrapolation, to intended clinical doses. These studies should also include experimental groups which will receive doses significantly higher than proposed clinical doses. Study duration should be at least equal to the proposed length of the clinical trials. For example, if the clinical trial duration is proposed to be less than two weeks, the length of preclinical studies should be at least two weeks. Finally, while vaginal products may be designed (and labelled) specifically for application to vaginal mucosa, the circumstances of product use are such that the product may come in contact with other mucosal surfaces. Also, active and/or inactive ingredients may be absorbed through the vaginal mucosa resulting in systemic exposure. Consequently, the safety assessment of new vaginal products should include oral repeat dose toxicity studies.

Use of a vaginal product for STD prevention may be a life-long practice, and as a consequence is considered to be a form of chronic drug administration. Therefore, before marketing a new vaginal product chronic dosing studies in animals are needed to identify potential problems with long term product use. In the latter phase of clinical development, either one year studies in two animal species (one non-rodent), or two year rodent carcinogenicity bioassays, should be conducted. In repeat-dose studies, determinations of pharmacokinetic parameters should be incorporated, especially to determine the extent of systemic exposure to the drug. Special toxicology studies should also be performed, including determinations of the potential for mutagenicity, irritation, hypersensitization and photosensitization. Bearing in mind the site of administration, a special emphasis should be placed on conducting preclinical reproductive toxicology studies as early as possible in the clinical development of the drug. A more detailed presentation of preclinical safety requirements is available from the FDA in the form of a "Points to Consider" document developed by toxicologists in the Division of Antiviral Drug Products (DAVDP), Center for Drug Evaluation and Research. This informal guidance document can be obtained through the DAVDP preIND program¹⁵.

Chemistry, Manufacturing and Controls

A common misconception about the regulatory process is the belief that the FDA approves "drugs" (i.e., a given chemical substance). In point of fact, the agency approves a specific product for a given indication, manufactured in a reproducible way at a particular site (or sites), according to very well defined specifications. Issues related to product composition, stability, reproducibility and the manufacturing process are addressed in the Chemistry, Manufacturing and Controls (CMC) section of an NDA The accumulation of information for this section of the NDA begins at the preIND stage when the drug substance (i.e., the active ingredient) is produced and characterized. Information gathering continues throughout IND development as the final formula is created and optimized, and controls are established. By Phase 2 of the IND the final formula should be set for clinical trials and stability studies on the packaged drug product should be in progress.

The CMC section of the NDA deals with every aspect of product manufacture, from the isolation, characterization, purification and quantitation of the active ingredient(s) through complete formulation and manufacture of the final product. This includes the establishment of definitive specifications for each step in the production process, including quality control procedures and guidelines that must be met for content, purity, reproducibility (i.e., lot-to-lot variation), stability and bioavailability. Many of these CMC requirements are defined by law and are described in detail in the Code of Federal Regulations. However, as noted for the preclinical toxicology requirements, informal guidance documents regarding CMC issues in product development can be obtained through the DAVDP preIND program.

It is important that drug activity, toxicity and CMC studies not be conducted in a vacuum during preclinical and early clinical drug development. The clinical development plan for a vaginal WV prophylaxis product should be supported by preclinical data on activity and toxicity generated with the actual product that will be tested in the clinic. However, the manner in which a drug is formulated for human use can have a significant impact on the activity and/or toxicity of the final product. It is clear from previous contraception and STD prevention studies that even minor modifications in formulation, including changes in presumably "inactive ingredients", can alter the efficacy and/or toxicity of a vaginal preparation. For example, foaming products appear to be inherently more irritating than formulations prepared with nonfoaming excipients¹⁶. In contrast, some "inactive" ingredients can have a soothing or protective effect on the mucosa. Recognition of this potential has prompted research on new delivery methods and excipients that can shield vaginal and cervical surfaces from the direct cytotoxic activity of detergents like N-9 without compromising contraceptive or antimicrobial activity. It is important to note that the relative clinical importance of these changes in toxicity and/or activity may vary with intended use. For a contraceptive product, relatively small changes in irritancy potential may influence compliance, but are otherwise unlikely to affect efficacy. In contrast, disruption of the vaginal mucosa is believed to facilitate HIV transmission, potentially impacting on the ability of a product to protect susceptible host cells from infection. Indeed, conflicting clinical data on the efficacy of N-9 products for HIV prevention may reflect differences in the level of mucosal irritation induced by the different products

 

In addition to direct cytotoxicity, changes in formulation can also influence other factors that affect product performance in the clinic. Allergenicity is an important concern, both in terms of direct and indirect induction of an allergic event. Another important form of toxicity that may be affected by changes in the type or quantity of active or inactive ingredients is disruption of the normal vaginal ecosystem. Finally, other product characteristics such as distribution and retention in the vaginal vault, duration of activity in situ, systemic absorption, bioavailability of the active ingredient(s), and miscibility with cervico-vaginal secretions are all subject to changes in formulation. Since each of these characteristics can influence the utility of a given product, it is important to give consideration to these issues when conducting formulation studies.

Clinical Testing

Adequate and well controlled clinical trials form the cornerstone of the NDA process, particularly for products seeking an STD prevention indication where the local environment may have a substantial impact on efficacy. Many compounds that have antiviral activity in vitro will prove to be unsuitable for human use due to factors such as direct or indirect toxicity) poor bioavail ability, formulation instability, or a loss of activity following exposure to body fluids. While it is possible to identify some of these problems via nonclinical studies, it is often difficult to accurately predict the ultimate impact on product performance in humans, As a consequence, data from clinical efficacy trials are needed to demonstrate that a product can function successfully within the vaginal environment to prevent the sexual transmission of HIV. Clinical data may also provide essential information about "real life" limitations of product activity such as duration of activity and dosing requirements.

The data from controlled clinical trials conducted with a new vaginal product will be analyzed to assess both safety and efficacy. This generally involves a subject by subject review by the FDA of the data from at least two independent, preferably prospective studies with concurrent controls. These trials need not be identical, in fact, they may be designed to answer slightly different questions or they may target different risk groups. The goal of the clinical development program is to provide two independent assessments of activity. It should be noted that while safety and efficacy data may be obtained from the same studies, not all data sets can be used to establish efficacy. NDA efficacy standards do not permit the use of anecdotal data or testimonials to demonstrate activity. In contrast, the safety assessment will consider all sources of information. The purpose of using an expanded database for safety determinations is to try to gain as much information as possible about the potential for adverse experiences associated with product use prior to widespread marketing.

This product category (i.e., topical STD) prophylaxis), is particularly challenging with respect to clinical trial design and implementation. A thorough discussion of the difficulties inherent in the conduct of WV prevention trials is beyond the scope of this chapter. However, a potential sponsor should be aware of a number of ethical and practical issues that are relevant to the clinical assessment of new drug or biologic-based HIV prevention strategies. Given that HIV is a life-threatening disease, ethical considerations mandate that participants in these trials be educated about risk factors for acquiring HIV infection and encouraged to use condoms during intercourse. These interventions alone will impact on the rate of seroconversion in most cohorts and will need to be taken into account when sample size estimates are made.

Compliance is another important factor that should be addressed, not only to support efficacy evaluations but also to facilitate accuracy in product labelling. Unlike conventional drug treatment trials where an "intent to treat" analysis is routinely used, contraceptive products are described in terms of "perfect use effectiveness", that is the degree of protection that can be anticipated if care is taken to use the product consistently and correctly. A similar approach is needed for vaginal products intended to prevent sexually transmitted diseases, irrespective of their contraceptive activity. It is important to recognize that the degree to which compliance can be determined and the methods that can be used to measure it will vary with the population under study

Selection of the target population is also key issue in trial design, particularly with respect to the ability to generalize the results to other risk groups. Commercial sex workers and patients in STD clinics at high risk for acquiring HIV infection are likely to differ from more low risk populations in a number of important respects. The frequency of product use and thus, the potential for product related toxicity, the relative impact of educational intervention, the presence of other risk factors that may facilitate HIV transmission (e.g., ulcerative STDs), and compliance may differ substantially between these cohorts. Each of these factors can, in turn, affect product efficacy. For example, the presence of other STDs that can influence the rate of HIV infection in a high risk cohort could overestimate product activity in a lower risk population, or underestimate it. If a product worked extremely well in frequent users with little or no toxicity, it is reasonable to assume that it would be safe and effective in a lower risk population. However, if the product exerted its protective effect exclusively through prevention of other STDs that facilitate HIV transmission, efficacy may not be as great in individuals at low risk for these other diseases. In contrast, the presence of ulcerative STDs may require that a product have very rapid, high level antiviral activity in order to provide protection against HIV, transmission. This high degree of potency may not be necessary to protect low risk individuals with a normal vaginal ecosystem from infection.

Clearly, formulating a scientifically sound and ethically acceptable clinical trial design for an vaginal STD prophylaxis product can be challenging. In addition to the issues discussed above, the selection of appropriate endpoints, methods for monitoring toxicity, and the types of controls that should be used are also important factors to consider. Recognizing the difficulties associated with the development of new vaginal products, a number of government-based and independent groups have initiated efforts to facilitate research in this area, The Interagency Working Group on Vaginal Microbicides (IAWG), organized by the World Health Organization (WHO), prepared a master protocol can that be used as a guide for the design of clinical safety and efficacy trials for new vaginal products seeking an STD prevention indication²⁰. Sponsors of promising new agents can also seek help from groups at the NIH where there are programs in place to assist with clinical, and in some cases, preclinical testing. The Division of Antiviral Drug Products at the FDA provides guidance early in product development through the preIND program, then continues to work with sponsors on a more formal basis throughout the IND and NDA process. For products that also have contraceptive activity, a "Points to Consider" document pertaining to the development of new vaginal contraceptives is available from the Division of Metabolism and Endocrine Drug Products at the FDA ²¹.

With the rapid rise in heterosexual transmission of HIV, the need for female-controlled methods to prevent infection has become a critical factor in public health efforts to control this epidemic. In the absence of a safe and effective prophylactic vaccine, a variety of different types of products, including contraceptive and non-contraceptive agents, will be needed to meet the needs of women at risk for acquiring this infection. The FDA recognizes the important role that these products can play in efforts to control the HIV epidemic and encourages sponsors of promising agents to seek input from the agency early in the product development process. Good communication and a productive working relationship can help to expedite the development and approval of new products that can stop the spread of HIV infection in this country and eventually, worldwide.

References:

1. Global Programme on AIDS. "The current global situation of the MIV/AIDS pandemic. July Press Release, 1992.

2. Centers for Disease Control and Prevention. Update: Acquired Immunodeficiency Syndrome - United States, 1992 MMWR 42:547, 1993.

3. CDC. HIV/AIDS surveillance report. Atlanta: US Department of Health and Human Services, Public Health Service, 6(#1):20,25-7 (1994).

4. CDC. Heterosexually acquired AIDS--United States, 1993 MMWR 43:155-60(1994).

5. European Study Group on Heterosexual Transmission of "IV. Comparison of female-to-male and male-to female transmission of HIV in 563 stable couples. Br Med J; 304:809-13 (1992).

6. Saracco A, et al. Man-to-woman sexual transmission of HIV: Longitudinal study of 343 steady partners of infected men. J Acquir Immun Defic Syndr 5: 497-502 (1993).

7. Worth, D. Sexual decision making and AIDS: why condom promotion among vulnerable women is likely to fail. Stud Fam Plann 10:297-307 (1989).

8. Ulin, PR. African women and AIDS: negotiating behavioral change. Soc Sci Med 34:63-73 (1992).

9. Stone KM, Grimes DA, Magder LS. Personal protection against sexually transmitted diseases. Am J Obstet Gynecol 155:180-88 (1986).

10. Stein, ZA. HIV prevention: the need for methods women can use. Am S Public Health 80:460-462 (1990).

11. Department of Health and Human Services; Food and Drug Administration: Vaginal contraceptive drug products for over-the-counter human use; proposed rule making. Federal Register 60, #23: 6892(1995).

12. Levy, J. Pathogenesis of Human Immunodeficiency Virus Infection. Microbiological Reviews 57,#1:183-289 (1993).

13. Bon-Habib, D. et al: Cryptic nature of envelope V3 region epitopes protects primary monocytotropic human immunodeficiency virus type 1 from antibody neutralization.J. Virology 68(9): 6006-6013 (1994).

14. Oravecz, T. et al. CD26 correlates with entry, replication and cytopathicity of monocytotropic HIV-I strains in a T-cell line. Nature Medicine 1(9):919-926 (1995).

15 "Points to Consider in the Preclinical Pharmacology/Toxicology Development of Topical Microbicides Intended for Prevention of Viral Sexually Transmitted Diseases Including HIV. Informal communication available from the Division of Antiviral Drug Products, Center for Drug Evaluation and Research, U.S. Food and Drug Administration. 5600 Fishers Lane, Rockville, MD 20857, 301-443-9563. Attention: Consumer Safety Officer for preIND Affairs.

16. Goeman et al. Frequent use of Menfegol spermicidal vaginal foaming tablets associated with a high incidence of genital lesions. JID 171:1611-1614 (1995).

17. Kreiss J. et al. Efficacy of nonoxynol-9 contraceptive sponge use in preventing heterosexual acquisition of HIV in Nairobi prostitutes. JAMA 268:477-482 (1992).

18. Zekeng et al. Barrier contraceptive use and HIV infection among high-risk women in Cameroon. AIDS 7:725-731 (1993).

19. Roddy RE, et al. Dosing study of nonoxynol-9 and genital irritation. Inter J STD & AIDS 4:165-170 (1993).

20. Clinical Testing of Vaginal Microbicides: Guidelines for Safety and Efficacy Studies. A current draft version of this document can be obtained from the WHO, the NIH (NICHD or NIAID) or the FDA (Division of Antiviral Drug Products, CDER).

21. "Guidance for development of Vaginal Contraceptive Drugs." Informal communication from the Division of Metabolism and Endocrine Drug Products, Center for Drug Evaluation and Research, U.S. Food and Drug Administration. 5600 Fishers Lane, Rockville. MD 20857.

Note: All ODE IV Pre-IND communications are considered informal under 21 CFR 10.90(b)(9) and represent the best judgment of the Office of Drug Evaluation IV at this time. The communications do not necessarily represent the formal position of the Center for Drug Evaluation and Research or the Food and Drug Administration, and does not bind or otherwise obligate the Center or Agency to the views expressed. Please direct any comments regarding ODE IV Pre-IND/Topical Microbicide Web Pages to ode4preind@cder.fda.gov